Optimizing Supply Chain with Time Series Forecasting: A Customer-Centric Approach

Author(s): Shenggang Li

Originally published on Towards AI.

Using the Repurchase Predictive Model for Product Demand Forecasting

This member-only story is on us. Upgrade to access all of Medium.

In this paper, I explore a time series forecasting model to predict demand in the supply chain industry. I aim to forecast sales for multiple products over the next N days.

I tried several traditional methods, including ARIMA and Prophet, but found that these statistical models weren’t quite suitable given the complexity and diversity of the products. As a result, I sought a better approach.

That’s when I thought of a different strategy — using the Lifetimes Python library to predict customer behavior at a granular level.

To complete the demand forecasting, I will use transaction data that differs from traditional time series data.

A key characteristic of this time series data is that it is based on transactions involving the customer, product, and time, rather than just product and time as in traditional time series data.

Even though we are predicting the same outcome — the sales volume of a product over a future period — differences in data structures and business logic may require us to use different models or methods for better predictions and explanations.

For instance, in this project, which involves a typical supply chain… Read the full blog for free on Medium.

Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.

Published via Towards AI